A CNN algorithm for classification of possible melanoma diagnosis using skin lesion images. The training data for lesion diagnosis which consists of 10015 images are downloaded from ISIC 2018: Skin Lesion Analysis towards Melanoma Detection”. There are seven possible disease categories: Melanoma, Melanocytic Nevus, Basal Cell Carcinoma, Actinic Keratosis, Benign Keratosis Dermatofibroma and Vascular Lesion.
Created a convolutional neural network for melanoma classification using some of the ideas fundamental to LeNet, which was applied to recognize hand-written characters. Considering the fact that we need a deeper network for this task,I added more layers and the final version of the network consists of 7 layers The network has four convolutional and three pooling layers for feature extraction, and one fully-connected layers, in the end for classification. This contribution improves the performance of the network, despite having some additional features which increase the training time.